I am running windows 10, core i7-8700 cpu, gtx geforce 1660 ti GPU.
When training models, gpu utilization is very low (5-10% at max, sometimes lower).
Even is network is five layers. CPU utilization on the other hand is 30% and above.
Please check the following:
The CUDA and CuDNN versions match. According to the statistics, it may very well use the CPU instead of GPU while training. You can try to see if your GPU is available below option 2.
If the former is solved, you may want to increase the batch_size, in case there is a very small batch size. It may be the case that TensorFlow pre-allocates a small amount of GPU to your training.
For step 1, in order to verify that the video card is both available and used, make use of the next lines of code:
import tensorflow as tf
print("Num GPUs Available: ", len(tf.config.experimental.list_physical_devices('GPU')))
tf.debugging.set_log_device_placement(True)
# Create some tensors
a = tf.constant([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]])
b = tf.constant([[1.0, 2.0], [3.0, 4.0], [5.0, 6.0]])
c = tf.matmul(a, b)
print(c)
The print should contain(along the result) the following information:
Executing op MatMul in device
/job:localhost/replica:0/task:0/device:GPU:0
Related
I'm doing some hyper-parameter tuning, so speed is key. I've got a nice workstation with both an AMD Ryzen 9 5950x and an NVIDIA RTX3060ti 8GB.
Setup:
xgboost 1.5.1 using PyPi in an anaconda environment.
NVIDIA graphics driver 471.68
CUDA 11.0
When training a xgboost model using the scikit-learn API I pass the tree_method = gpu_hist parameter. And i notice that it is consistently outperformed by using the default tree_method = hist.
Somewhat surprisingly, even when I open multiple consoles (I work in spyder) and start an Optuna study in each of them, each using a different scikit-learn model until my CPU usage is at 100%. When I then compare the tree_method = gpu_hist with tree_method = hist, the tree_method = hist is still faster!
How is this possible? Do I have my drivers configured incorrectly?, is my dataset too small to enjoy a benefit from the tree_method = gpu_hist? (7000 samples, 50 features on a 3 class classification problem). Or is the RTX3060ti simply outclassed by the AMD Ryzen 9 5950x? Or none of the above?
Any help is highly appreciated :)
Edit #Ferdy:
I carried out this little experiment:
def fit_10_times(tree_method, X_train, y_train):
times = []
for i in range(10):
model = XGBClassifier(tree_method = tree_method)
start = time.time()
model.fit(X_train, y_train)
times.append(time.time()-start)
return times
cpu_times = fit_10_times('hist', X_train, y_train)
gpu_times = fit_10_times('gpu_hist', X_train, y_train)
print(X_train.describe())
print('mean cpu training times: ', np.mean(cpu_times), 'standard deviation :',np.std(cpu_times))
print('all training times :', cpu_times)
print('----------------------------------')
print('mean gpu training times: ', np.mean(gpu_times), 'standard deviation :',np.std(gpu_times))
print('all training times :', gpu_times)
Which yielded this output:
mean cpu training times: 0.5646213531494141 standard deviation : 0.010005875058323703
all training times : [0.5690040588378906, 0.5500047206878662, 0.5700047016143799, 0.563004732131958, 0.5570034980773926, 0.5486617088317871, 0.5630037784576416, 0.5680046081542969, 0.57651686668396, 0.5810048580169678]
----------------------------------
mean gpu training times: 2.0273998022079467 standard deviation : 0.05105794761358874
all training times : [2.0265607833862305, 2.0070691108703613, 1.9900789260864258, 1.9856727123260498, 1.9925382137298584, 2.0021069049835205, 2.1197071075439453, 2.1220884323120117, 2.0516715049743652, 1.9765043258666992]
The peak in CPU usage refers to the CPU training runs, and the peak in GPU usage the GPU training runs.
7000 samples is too small to fill the GPU pipeline, your GPU is likely to be starving. We usually work with millions of samples when using GPU acceleration.
I am trying to run StyleGAN2 using a cluster equipped with eight GPUs (NVIDIA GeForce RTX 2080). At present, I am using the following configuration in training_loop.py:
minibatch_size_dict = {4: 512, 8: 256, 16: 128, 32: 64, 64: 32}, # Resolution-specific overrides.
minibatch_gpu_base = 8, # Number of samples processed at a time by one GPU.
minibatch_gpu_dict = {}, # Resolution-specific overrides.
G_lrate_base = 0.001, # Learning rate for the generator.
G_lrate_dict = {}, # Resolution-specific overrides.
D_lrate_base = 0.001, # Learning rate for the discriminator.
D_lrate_dict = {}, # Resolution-specific overrides.
lrate_rampup_kimg = 0, # Duration of learning rate ramp-up.
tick_kimg_base = 4, # Default interval of progress snapshots.
tick_kimg_dict = {4:10, 8:10, 16:10, 32:10, 64:10, 128:8, 256:6, 512:4}): # Resolution-specific overrides.
I am training using a set of 512x52 pixel images. After a couple of iterations, I get the error message reported below and it looks like the script stops running (using watch nvidia-smi, we have that both the temperature and the fan activity for the GPUs decreases). I already reduced the batch size but it looks like the problem is somewhere else. Do you have any tip on how to fix this?
I was able to run StyleGAN with the same dataset. In the paper they say that StyleGAN2 should be less heavy, so I am a bit surprised.
Here is the error message I get:
2019-12-16 18:22:54.909009: E tensorflow/stream_executor/cuda/cuda_driver.cc:828] failed to allocate 334.11M (350338048 bytes) from device: CUDA_ERROR_OUT_OF_MEMORY: out of memory
2019-12-16 18:22:54.909087: W tensorflow/core/common_runtime/bfc_allocator.cc:314] Allocator (GPU_0_bfc) ran out of memory trying to allocate 129.00MiB (rounded to 135268352). Current allocation summary follows.
2019-12-16 18:22:54.918750: W tensorflow/core/common_runtime/bfc_allocator.cc:319] **_***************************_*****x****x******xx***_******************************_***************
2019-12-16 18:22:54.918808: W tensorflow/core/framework/op_kernel.cc:1502] OP_REQUIRES failed at conv_grad_input_ops.cc:903 : Resource exhausted: OOM when allocating tensor with shape[4,128,257,257] and type float on /job:localhost/replica:0/task:0/device:GPU:0 by allocator GPU_0_bfc
The config-f model for StyleGAN2 is actually bigger than StyleGAN1. Try using a less VRAM consuming configuration like config-e. You can actually change the configuration of the model by passing a flag in your python command like so: https://github.com/NVlabs/stylegan2/blob/master/run_training.py#L144
In my case, I'm able to train StyleGAN2 with config-e on 2 RTX 2080ti.
One or more high-end NVIDIA GPUs, NVIDIA drivers, CUDA 10.0 toolkit
and cuDNN 7.5. To reproduce the results reported in the paper, you
need an NVIDIA GPU with at least 16 GB of DRAM.
Your NVIDIA GeForce RTX 2080 card has 11GB, but I guess you're saying you have 8 of them? I don't think tensorflow is setup for parallelism out of the box.
I'd tried dozen times to install tensorflow-GPU, but all failed, thougth I followed the instructions indicated by those who successfully installed.
I just noticed that I have 2 GPU indicators, meaning I have 2 GUP set? One Intel and one Nvidia.
Could it be the reason? If so, how can I kick in the proper one (Nvidia)?
Thanks for advice.
Verify proper installation of tensorflow-GPU by using following steps,
Can you check if tensorflow is using GPU,
import tensorflow as tf
print(tf.test.is_gpu_available())
To find out which device is used, you can enable log device placement
like this:
sess = tf.Session(config=tf.ConfigProto(log_device_placement=True))
you can try to assign a computation to the gpu and see whether you have an error
import tensorflow as tf
with tf.device('/gpu:0'):
a = tf.constant([1.0, 2.0, 3.0, 4.0, 5.0, 6.0], shape=[2, 3], name='a')
b = tf.constant([1.0, 2.0, 3.0, 4.0, 5.0, 6.0], shape=[3, 2], name='b')
c = tf.matmul(a, b)
with tf.Session() as sess:
print (sess.run(c))
Check the GPU usage using
nvidia-smi
Running AWS SageMaker with a custom model, the TrainingJob fails with an Algorithm Error when using Keras plus a Tensorflow backend in multi-gpu configuration:
from keras.utils import multi_gpu_model
parallel_model = multi_gpu_model(model, gpus=K)
parallel_model.compile(loss='categorical_crossentropy',
optimizer='rmsprop')
parallel_model.fit(x, y, epochs=20, batch_size=256)
This simple parallel model loading will fail. There is no further error or exception from CloudWatch logging. This configuration works properly on local machine with 2x NVIDIA GTX 1080, same Keras Tensorflow backend.
According to SageMaker documentation and tutorials the multi_gpu_model utility will work ok when Keras backend is MXNet, but I did not find any mention when the backend is Tensorflow with the same multi gpu configuration.
[UPDATE]
I have updated the code with the suggested answer below, and I'm adding some logging before the TrainingJob hangs
This logging repeats twice
2018-11-27 10:02:49.878414: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1511] Adding visible gpu devices: 0, 1, 2, 3
2018-11-27 10:02:49.878462: I tensorflow/core/common_runtime/gpu/gpu_device.cc:982] Device interconnect StreamExecutor with strength 1 edge matrix:
2018-11-27 10:02:49.878471: I tensorflow/core/common_runtime/gpu/gpu_device.cc:988] 0 1 2 3
2018-11-27 10:02:49.878477: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1001] 0: N Y Y Y
2018-11-27 10:02:49.878481: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1001] 1: Y N Y Y
2018-11-27 10:02:49.878486: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1001] 2: Y Y N Y
2018-11-27 10:02:49.878492: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1001] 3: Y Y Y N
2018-11-27 10:02:49.879340: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] Created TensorFlow device (/device:GPU:0 with 14874 MB memory) -> physical GPU (device: 0, name: Tesla V100-SXM2-16GB, pci bus id: 0000:00:1b.0, compute capability: 7.0)
2018-11-27 10:02:49.879486: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] Created TensorFlow device (/device:GPU:1 with 14874 MB memory) -> physical GPU (device: 1, name: Tesla V100-SXM2-16GB, pci bus id: 0000:00:1c.0, compute capability: 7.0)
2018-11-27 10:02:49.879694: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] Created TensorFlow device (/device:GPU:2 with 14874 MB memory) -> physical GPU (device: 2, name: Tesla V100-SXM2-16GB, pci bus id: 0000:00:1d.0, compute capability: 7.0)
2018-11-27 10:02:49.879872: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] Created TensorFlow device (/device:GPU:3 with 14874 MB memory) -> physical GPU (device: 3, name: Tesla V100-SXM2-16GB, pci bus id: 0000:00:1e.0, compute capability: 7.0)
Before there is some logging info about each GPU, that repeats 4 times
2018-11-27 10:02:46.447639: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1432] Found device 3 with properties:
name: Tesla V100-SXM2-16GB major: 7 minor: 0 memoryClockRate(GHz): 1.53
pciBusID: 0000:00:1e.0
totalMemory: 15.78GiB freeMemory: 15.37GiB
According to the logging all the 4 GPUs are visible and loaded in the Tensorflow Keras backend. After that no application logging follows, the TrainingJob status is inProgress for a while, after that it becomes Failed with the same Algorithm Error.
Looking at CloudWatch logging I can see some metrics at work. Specifically GPU Memory Utilization, CPU Utilization are ok, while GPU utilization is 0%.
[UPDATE]
Due to a known bug on Keras that is about saving a multi gpu model, I'm using this override of the multi_gpu_model utility in keras.utils
from keras.layers import Lambda, concatenate
from keras import Model
import tensorflow as tf
def multi_gpu_model(model, gpus):
#source: https://github.com/keras-team/keras/issues/8123#issuecomment-354857044
if isinstance(gpus, (list, tuple)):
num_gpus = len(gpus)
target_gpu_ids = gpus
else:
num_gpus = gpus
target_gpu_ids = range(num_gpus)
def get_slice(data, i, parts):
shape = tf.shape(data)
batch_size = shape[:1]
input_shape = shape[1:]
step = batch_size // parts
if i == num_gpus - 1:
size = batch_size - step * i
else:
size = step
size = tf.concat([size, input_shape], axis=0)
stride = tf.concat([step, input_shape * 0], axis=0)
start = stride * i
return tf.slice(data, start, size)
all_outputs = []
for i in range(len(model.outputs)):
all_outputs.append([])
# Place a copy of the model on each GPU,
# each getting a slice of the inputs.
for i, gpu_id in enumerate(target_gpu_ids):
with tf.device('/gpu:%d' % gpu_id):
with tf.name_scope('replica_%d' % gpu_id):
inputs = []
# Retrieve a slice of the input.
for x in model.inputs:
input_shape = tuple(x.get_shape().as_list())[1:]
slice_i = Lambda(get_slice,
output_shape=input_shape,
arguments={'i': i,
'parts': num_gpus})(x)
inputs.append(slice_i)
# Apply model on slice
# (creating a model replica on the target device).
outputs = model(inputs)
if not isinstance(outputs, list):
outputs = [outputs]
# Save the outputs for merging back together later.
for o in range(len(outputs)):
all_outputs[o].append(outputs[o])
# Merge outputs on CPU.
with tf.device('/cpu:0'):
merged = []
for name, outputs in zip(model.output_names, all_outputs):
merged.append(concatenate(outputs,
axis=0, name=name))
return Model(model.inputs, merged)
This works ok on local 2x NVIDIA GTX 1080 / Intel Xeon / Ubuntu 16.04. It will fails on SageMaker Training Job.
I have posted this issue on AWS Sagemaker forum in
TrainingJob custom algorithm with Keras backend and multi GPU
SageMaker Fails when using Multi-GPU with
keras.utils.multi_gpu_model
[UPDATE]
I have slightly modified the tf.session code adding some initializers
with tf.Session() as session:
K.set_session(session)
session.run(tf.global_variables_initializer())
session.run(tf.tables_initializer())
and now at least I can see that one GPU (I assume device gpu:0) is used from the instance metrics. The multi-gpu does not work anyways.
This might not be the best answer for your problem, but this is what I am using for a multi-gpu model with Tensorflow backend. First i initialize using:
def setup_multi_gpus():
"""
Setup multi GPU usage
Example usage:
model = Sequential()
...
multi_model = multi_gpu_model(model, gpus=num_gpu)
multi_model.fit()
About memory usage:
https://stackoverflow.com/questions/34199233/how-to-prevent-tensorflow-from-allocating-the-totality-of-a-gpu-memory
"""
import tensorflow as tf
from keras.utils.training_utils import multi_gpu_model
from tensorflow.python.client import device_lib
# IMPORTANT: Tells tf to not occupy a specific amount of memory
from keras.backend.tensorflow_backend import set_session
config = tf.ConfigProto()
config.gpu_options.allow_growth = True # dynamically grow the memory used on the GPU
sess = tf.Session(config=config)
set_session(sess) # set this TensorFlow session as the default session for Keras.
# getting the number of GPUs
def get_available_gpus():
local_device_protos = device_lib.list_local_devices()
return [x.name for x in local_device_protos if x.device_type == 'GPU']
num_gpu = len(get_available_gpus())
print('Amount of GPUs available: %s' % num_gpu)
return num_gpu
Then i call
# Setup multi GPU usage
num_gpu = setup_multi_gpus()
and create a model.
...
After which you're able to make it a multi GPU model.
multi_model = multi_gpu_model(model, gpus=num_gpu)
multi_model.compile...
multi_model.fit...
The only thing here that is different from what you are doing is the way Tensorflow is initializing the GPU's. I can't imagine it being the problem, but it might be worth trying out.
Good luck!
Edit: I noticed sequence to sequence not being able to work with multi GPU. Is that the type of model you are trying to train?
I apologize for the slow response.
It seems there are a lot of threads that are running in parallel, and I want to link them together, so that other individuals who have the same issue can see the progress and discussion going on.
https://forums.aws.amazon.com/thread.jspa?messageID=881541
https://forums.aws.amazon.com/thread.jspa?messageID=881540
https://github.com/aws/sagemaker-python-sdk/issues/512
There a few questions in regards to this.
What version of TensorFlow and Keras?
I am not too sure what is causing this problem. Does your container have all of the needed dependencies such as CUDA and etc? https://www.tensorflow.org/install/gpu
Were you able to train using single GPU with Keras?
I installed tensorflow with gpu, cuda 7.0 and cudnn 6.5. When I import tensorflow it works well.
I am trying to run a simple matrix multiplication on Tensorflow and it doesn't want to use my gpu though it seems to recognize it. I have this issue on my computer with a nvidia geforce 970m and on a cluster with two titan Z.
My first code is :
import tensorflow as tf
import numpy as np
size=100
#I create 2 matrix
mat1 = np.random.random_sample([size, size])*100
mat2 = np.random.random_sample([size, size])*100
a = tf.constant(mat1)
b = tf.constant(mat2)
c = tf.matmul(a, b)
sess = tf.Session(config=tf.ConfigProto(log_device_placement=True))
sess.run(c)
This code works and the result is :
Const_1: /job:localhost/replica:0/task:0/gpu:0
I tensorflow/core/common_runtime/simple_placer.cc:289] Const_1: /job:localhost/replica:0/task:0/gpu:0
Const: /job:localhost/replica:0/task:0/gpu:0
I tensorflow/core/common_runtime/simple_placer.cc:289] Const: /job:localhost/replica:0/task:0/gpu:0
MatMul: /job:localhost/replica:0/task:0/cpu:0
I tensorflow/core/common_runtime/simple_placer.cc:289] MatMul: /job:localhost/replica:0/task:0/cpu:0
So in my way, tensorflow uses my gpu to create constant but not for matmul (that is weird). Then, I force the gpu like this :
with tf.device("/gpu:0"):
a = tf.constant(mat1)
b = tf.constant(mat2)
c = tf.matmul(a, b)
sess = tf.Session(config=tf.ConfigProto(log_device_placement=True))
sess.run(c)
And Tensorflow returns :
InvalidArgumentError: Cannot assign a device to node 'MatMul': Could not satisfy explicit device specification '/gpu:0'
If someone have the same problem or an idea, I will be glad to read your answer !
I do not have enough reputation to comment, I have come across a similar issue, my question is here
TensorFlow: critical graph operations assigned to cpu rather than gpu